PROJECT SUMMARY Risk for most human diseases is attributable to segregating alleles at many interacting genes with environmentally sensitive effects. Future developments towards personalized precision medicine require a predictive understanding of how DNA sequence variants give rise to phenotypic variation through modulation of regulatory gene networks. This is challenging in human populations because variants associated with complex traits are embedded in relatively large local linkage disequilibrium (LD) blocks, within which segregating molecular polymorphisms are not independent. Thus, these variants are not necessarily causal, but could be in LD with the true common or rare causal variant(s) within the same LD block. Furthermore, the majority of variants associated with complex traits are in intergenic regions, up- or down-stream of coding regions, or in introns. These variants are presumably regulatory and affect variation in gene expression. Formally proving the causal relationships between molecular genetic variation, genetic variation in gene expression and other intermediate molecular phenotypes, and genetic variation in quantitative trait phenotypes is not possible in human populations. The Drosophila melanogaster Genetic Reference Panel (DGRP) was generated in our laboratories and consists of 205 inbred, sequenced lines derived from single inseminated females collected from the Raleigh, NC Farmer?s Market. We have used the DGRP to perform genome wide association (GWA) mapping for many organismal quantitative traits as well as genome wide gene expression, which has generated testable hypotheses about the genotype-phenotype map, including sex-, genetic background- and environment-specific effects. The precision of GWA mapping in the DGRP is excellent because of rapid local decline of LD with physical distance. Here, we propose to test these hypotheses using CRISPR/Cas9 mediated precise allelic replacement to functionally validate (1) additive, epistatic and environment-specific effects of common variants that affect chill coma recovery time; (2) pleiotropic, epistatic and environment-specific effects of rare variants; and (3) novel transcribed regions (NTRs) and cis-trans transcriptional networks, and evaluate their effects on genome-wide expression and quantitative traits. These proposed studies will enable us to evaluate the direct and pleiotropic effects of common and rare variants, in both genic and intergenic regions, that are shared and distinct between males and females, both with respect to organismal quantitative trait phenotypes as well as genome wide gene expression. We will be able to explicitly evaluate the existence and magnitude of epistatic interactions for organismal phenotypes and gene expression traits and create ?designer? genotypes between epistatically interacting alleles in defined genetic backgrounds. These studies will greatly advance our understanding of how subtle naturally occurring molecular variation impacts gene expression and organismal phenotypes.